nbLooking back exactly one year ago, a certain part of the #Fediverse was involved with crucial data rescue operations. This was because the political data cleansing in the US started to unfold and a group of dedicated people chose to form a guerilla data rescue collective, which we now know as @SafeguardingResearch and #SciOp.
Today, I took the opportunity to take a look back at 2025's events while also covering present and future developments of this movement in the form of a lecture and hands-on within my #LIS studies course on e-Publishing.
It was such a relief to once again talk about positive movements in those dark times utilizing decentralized technologies for social impact, and to pay tribute to the marginalized groups who are keeping up their fights every day.
In the discussion afterwards, among Master students of Library and Information Sciences (most of us do also already work in libraries) we talked about how one can publish scientific works in order to prevent getting taken down due to ideological data cleansing.
We all agreed up on that FAIR #OpenAccess should be mandatory for all publications. I proposed a three-way approach for actually storing the publication data simultaneously: 1. on a repository of a trustworthy organization, 2. on a (own) publicly accessable website, and 3. on a repository powered by decentralized technologies of ones choice. I reflected up on a thread I read here back in April 2025, where among others @jonny and @nichtich had talked about this topic: https://neuromatch.social/@jonny/114310419885059486. Another take-away was to publish files as raw as possible, e.g. in plain text, Markdown or HTML. In addition to that, I would add that files (and metadata) should be stored tamper-resistant to ensure data integrity, and PDF files should be PDF/A of that kind that #DigiPres organizations currently recommend to be at least a bit more future-proof.
To take myself for an example, back a couple of years ago, starting with my bachelor thesis, I began to mirror all my publications to the decentralized storage network #IPFS, which also gives me data integrity and tamper resistance by content-addressing. Depending on the type of work I do also upload my stuff to my own website or Zenodo. I at least dual-store my work in raw text and PDF. The first setting or config I adjust with my word processing software is the PDF/A export setting, so that it will save files to PDF/A per default.
#SafeguardingResearch #DataRescue #DigitalPreservation #OpenScience #Science #GLAM #Libraries #DigitalHumanities #AcademicChatter #Research
Today, I took the opportunity to take a look back at 2025's events while also covering present and future developments of this movement in the form of a lecture and hands-on within my #LIS studies course on e-Publishing.
It was such a relief to once again talk about positive movements in those dark times utilizing decentralized technologies for social impact, and to pay tribute to the marginalized groups who are keeping up their fights every day.
In the discussion afterwards, among Master students of Library and Information Sciences (most of us do also already work in libraries) we talked about how one can publish scientific works in order to prevent getting taken down due to ideological data cleansing.
We all agreed up on that FAIR #OpenAccess should be mandatory for all publications. I proposed a three-way approach for actually storing the publication data simultaneously: 1. on a repository of a trustworthy organization, 2. on a (own) publicly accessable website, and 3. on a repository powered by decentralized technologies of ones choice. I reflected up on a thread I read here back in April 2025, where among others @jonny and @nichtich had talked about this topic: https://neuromatch.social/@jonny/114310419885059486. Another take-away was to publish files as raw as possible, e.g. in plain text, Markdown or HTML. In addition to that, I would add that files (and metadata) should be stored tamper-resistant to ensure data integrity, and PDF files should be PDF/A of that kind that #DigiPres organizations currently recommend to be at least a bit more future-proof.
To take myself for an example, back a couple of years ago, starting with my bachelor thesis, I began to mirror all my publications to the decentralized storage network #IPFS, which also gives me data integrity and tamper resistance by content-addressing. Depending on the type of work I do also upload my stuff to my own website or Zenodo. I at least dual-store my work in raw text and PDF. The first setting or config I adjust with my word processing software is the PDF/A export setting, so that it will save files to PDF/A per default.
#SafeguardingResearch #DataRescue #DigitalPreservation #OpenScience #Science #GLAM #Libraries #DigitalHumanities #AcademicChatter #Research
jonny (good kind) (@jonny@neuromatch.social)
to any academics who want their work to outlast their current hype cycle, the next publisher downsizing, or government book burning: - make your work available for free. - as a PDF or HTML document (or, fine, JATS) - from a URL that requires no login or javascript to access. having your work only available behind some heinous publisher paywall or 'enhanced reader' is a straightforward statement that you don't care about your work being part of the shared pursuit of human understanding. it is, in fact, your job to make sure that your work available. sincerely, someone trying to preserve your work.
Data Rescue Project #DataRescue
ResearchBuzz: Firehose
Paul R. Pival (he/him)
